Decision feedback equaliser design using support vector machines
IEE Proceedings - Vision Image and Signal ProcessingPublished 1 January 2000
Sheng Chen, Steve Gunn, C.J. Harris
Citations45
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A strategy is proposed for designing the DFE based on support vector machines (SVMs) that achieves asymptotically the MBER solution and is superior in performance to the usual MMSE solution.
Abstract
The conventional decision feedback equaliser (DFE) that employs a linear combination of channel observations and past decisions is considered. The design of this class of DFE is to construct a hyperplane that separates the different signal classes. It is well known that the popular minimum mean square error (MMSE) design is generally not the optimal minimum bit error rate (MBER) solution. A strategy is proposed for designing the DFE based on support vector machines (SVMs). The SVM design achieves asymptotically the MBER solution and is superior in performance to the usual MMSE solution. Unlike the exact MBER solution, this SVM solution can be computed very efficiently.
Keywords
Computer ScienceEngineering
The Nature of Statistical Learning Theory
39,279 Citations1995Vladimir Vapnik
Machine LearningSupport-Vector Networks
33,035 Citations1995Corinna Cortes, Vladimir Vapnik
High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Adaptive Filter Theory
12,680 Citations1986S. Haykin
This work focuses on Adaptive Filters Using Infinite-Duration Impulse Response Structures, a Stochastic Processes and Models that combines Kalman Filters as the Unifying Bases for RLS Filters with Estimation Theory and Complex Variables.
Computers & Operations ResearchNonlinear programming: theory and algorithms
5,829 Citations1993
The book is a solid reference for professionals as well as a useful text for students in the fields of operations research, management science, industrial engineering, applied mathematics, and also in engineering disciplines that deal with analytical optimization techniques.
Support Vector Machines for Classification and Regression
2,828 Citations1998S.R. Gunn
Support Vector Method for Function Approximation, Regression Estimation and Signal Processing
2,702 Citations1996Vladimir Vapnik, Steven E. Golowich +1 more
This presentation reports results of applying the Support Vector method to problems of estimating regressions, constructing multidimensional splines, and solving linear operator equations.
IEEE Communications MagazineAdaptive equalization
1,014 Citations1982Shahid Qureshi
A modulator-demodulator (MODEM) is required to carry digital signals over analog channels by translating binary data to voice-frequency signals and back by translating binary data to voice-frequency signals and back.
Journal of the Operational Research SocietyMathematical Programming: Theory and Algorithms.
570 Citations1987David K. Smith, Michel Minoux +1 more
IEEE Journal on Selected Areas in CommunicationsChannel equalization using adaptive complex radial basis function networks
218 Citations1995Inhyok Cha, S.A. Kassam
Results of simulations show that the CRBFN with the stochastic-gradient algorithm can be quite effective in channel equalization.
IEEE Transactions on Signal ProcessingAdaptive Bayesian equalizer with decision feedback
218 Citations1993Sheng Chen, B. Mulgrew +1 more
It is shown that the Bayesian equalizer has a structure equivalent to that of the radial basis function network, the latter being a one-hidden-layer artificial neural network widely used in pattern classification and many other areas of signal processing.
IEEE Transactions on Signal ProcessingThe application of nonlinear structures to the reconstruction of binary signals
179 Citations1991Gavin J. Gibson, S. Siu +1 more
The manner in which neural networks can be utilized as adaptive channel equalizers is described, and simulation results which suggest that the neural network equalizers offer a performance which exceeds that of the linear structures, particularly in the high-noise environment are presented.
Signal ProcessingComplex-valued radial basis function network, Part II: Application to digital communications channel equalisation
139 Citations1994Sheng Chen, Stephen McLaughlin +1 more
It is shown that the optimal Bayesian equaliser is structurally equivalent to the complex radial basis function network, and this intimate connection is exploited to develop fast training algorithms for implementing a Bayesianequaliser based on the latter.
IEE Proceedings I Communications Speech and VisionDecision feedback equalisation using neural network structures and performance comparison with standard architecture
121 Citations1990Gavin J. Gibson, S. Siu +1 more
Results indicate that the perceptron based decision feedbackequaliser provides better bit error rate performance relative to the least mean square decision feedback equaliser, especially in high noise conditions, and biterror rate performance degrades less owing to decision errors and is also less sensitive to gain variation.
IEEE Transactions on CommunicationsBlock decision feedback equalization
94 Citations1992D. Williamson, Rodney A. Kennedy +1 more
A natural generalization of the conventional decision feedback equalizer (DFE) based on block processing and maximum a posteriori decisions is presented.
IEEE Transactions on CommunicationsAdaptive Bayesian decision feedback equalizer for dispersive mobile radio channels
85 Citations1995Sheng Chen, Stephen McLaughlin +2 more
The paper develops a robust high performance Bayesian decision feedback equalizer (DFE) that outperforms the conventional DFE dramatically and the maximum likelihood sequence estimator (MLSE) for severely fading multipath channels.
IEE Proceedings - CommunicationsSpace translation properties and the minimum-BER linear-combiner DFE
61 Citations1998Sheng Chen, B. Mulgrew +2 more
It is demonstrated that the usual minimum mean square error (MMSE) solution does not achieve the full performance potential of the linear-combiner DFE structure, and a bit error rate (BER) expression is obtained and a method is proposed to optimally set the coefficients of thelinear-combiners DFE.
IEE Proceedings - CommunicationsBayesian decision feedback equaliser for overcoming co-channel interference
41 Citations1996Sheng Chen, Stephen McLaughlin +2 more
The results obtained indicate that, in the presence of severe co-channel interference, the Bayesian decision feedback equaliser employing the proposed simple scheme to compensate co-Channel interference can outperform the maximum likelihood sequence estimator that only treats co- channel interference as an additional coloured noise.
Approximate minimum bit-error rate equalization for binary signaling
37 Citations2002Chen-Chu Yeh, John R. Barry
This work proposes a simple stochastic gradient algorithm that approximately minimizes BER in the presence of linear intersymbol interference and white Gaussian noise and shows that the proposed algorithm compares favorably to the popular least-mean-square algorithm in terms of both steady-state performance and complexity.
IEEE Signal Processing MagazineEqualization techniques for high-density magnetic recording
28 Citations1998J.G. Proakis
It is observed that the optimum detector for the ISI-corrupted signal is a maximum-likelihood sequence detector (MLSD) that can be efficiently implemented by means of the Viterbi algorithm.
IEEE Signal Processing MagazineThe role of SP in data-storage systems
25 Citations1998Jaekyun Moon
This article addresses the issues relevant to signal processing in magnetic storage and discusses sequence detectors, with in-depth treatment given to those based on tree search, particularly finite or fixed delay tree search (FDTS) detectors.
IEE Proceedings - CommunicationsMinimum-SER linear-combiner decision feedback equaliser
21 Citations1999Sheng Chen, B. Mulgrew
The paper considers the conventional decision feedback equalizer (DFE) that employs a linear combination of the channel observations and the past decisions and derives an expression of the symbol error rate (SER) for the linear-combiner DFE with the general $M$-PAM constellation by utilizing a geometric translation property of decision feedback.
IEEE Transactions on CommunicationsA randomized bias technique for the importance sampling simulation of Bayesian equalizers
20 Citations1995R.A. Iltis
An importance sampling (IS) simulation technique is presented for Bayesian equalizers, based on the large deviations theory approach developed by Sadowsky and Bucklew, and results are presented for both recursive and non-recursive equalizers.